Scholarlink

<p>Hydrogen is widely produced via the Steam reforming of Methane (SMR) however, its effectiveness is heavily affected by non-linear and complex interdependencies between methane flow rate and reforming temperature. Conventional optimization techniques might not be able to determine global optimality, as increasing hydrogen production often increases the heat energy supplied to the furnace and convection-induced irreversibility’s, thus creating unexpected energy-exergy trade-offs. This study presents a novel multi-objective approach for SMR using energy-exergy modelling with the aid of Ant Colony Optimization (ACO) algorithm enhanced with Lévy flight to avoid premature convergence and local optima. In total 81 different datasets are developed using three levels of operation for each input variable to represent monotonic thermodynamic behavior. The Pearson correlation showed a strong relationship between hydrogen production and methane flow rate (<i>r</i> = 0.965, Adj. R² = 0.9303) and also between energy efficiency and reformer temperature (<i>r</i> = 0.88707, Adj. R² = 0.78419). A strong fitness function is developed with special focus on minimizing the convection and stack losses through the weighted priorities. The hybrid method found a 43% priority for energy efficiency, 35% priority for exergy efficiency, and 22% for hydrogen generation. Compared with standard ACO, which reached a fitness value of 0.86 in 55&#xa0;s, the proposed ACO–Lévy approach achieved a higher fitness value of 0.97 within 10.3&#xa0;s, corresponding to a 12.8% fitness improvement, 64.1% faster convergence, and 81.3% lower computational time. The suggested algorithm delivered an optimal outcome of 292 kmol/h of hydrogen with an energy efficiency of 74% and exergy efficiency of 70%, suggesting improved heat transfer and reduced irreversibility in the convection zone.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Energy–exergy optimization of steam methane reforming for hydrogen production using Lévy-flight-enhanced ant colony optimization

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月20日

查看全文
  • Md Shaquib Equbal,
  • Tasmeem Ahmad Khan,
  • Osama Khan,
  • Khaled Alnamasi,
  • Saad Arif,
  • Ashok Kumar Yadav

摘要

Hydrogen is widely produced via the Steam reforming of Methane (SMR) however, its effectiveness is heavily affected by non-linear and complex interdependencies between methane flow rate and reforming temperature. Conventional optimization techniques might not be able to determine global optimality, as increasing hydrogen production often increases the heat energy supplied to the furnace and convection-induced irreversibility’s, thus creating unexpected energy-exergy trade-offs. This study presents a novel multi-objective approach for SMR using energy-exergy modelling with the aid of Ant Colony Optimization (ACO) algorithm enhanced with Lévy flight to avoid premature convergence and local optima. In total 81 different datasets are developed using three levels of operation for each input variable to represent monotonic thermodynamic behavior. The Pearson correlation showed a strong relationship between hydrogen production and methane flow rate (r = 0.965, Adj. R² = 0.9303) and also between energy efficiency and reformer temperature (r = 0.88707, Adj. R² = 0.78419). A strong fitness function is developed with special focus on minimizing the convection and stack losses through the weighted priorities. The hybrid method found a 43% priority for energy efficiency, 35% priority for exergy efficiency, and 22% for hydrogen generation. Compared with standard ACO, which reached a fitness value of 0.86 in 55 s, the proposed ACO–Lévy approach achieved a higher fitness value of 0.97 within 10.3 s, corresponding to a 12.8% fitness improvement, 64.1% faster convergence, and 81.3% lower computational time. The suggested algorithm delivered an optimal outcome of 292 kmol/h of hydrogen with an energy efficiency of 74% and exergy efficiency of 70%, suggesting improved heat transfer and reduced irreversibility in the convection zone.